Our long-term programmatic goal is to utilize adult marrow-derived mesenchymal progenitors (mesenchymal stem cells - MSC) in tissue engineered constructs to repair craniofacial and long bone defects. MSC have robust potential to become osteoblasts and human MSC (hMSC) can form bone when implanted in heterotopic and orthotopic sites in animal models. A distinct advantage of exploiting the therapeutic potential of these stem cells is that autologous hMSC are easily cultivated from small marrow aspirates. Clinically, marrow could be collected from a patient, hMSC cultivated in vitro, and then delivered back to the patient. However, to most effectively use hMSC therapeutically, we must be able to fully control their differentiation. This requires a complete understanding of physiologically relevant factors that regulate hMSC differentiation, and in turn, downstream signaling and transcriptional networks that are activated. Our preliminary results show that bone morphogenetic-6 (BMP6) is an important autocrine mediator of hMSC differentiation. BMP6 is the only osteogenic BMP produced by hMSC and short-term treatment of cells with exogenous BMP6 promotes sustained expression of the osteogenic transcription factor osterix, leading to osteoblast differentiation. For this project we propose three specific aims: (1) we will examine the significance of endogenous BMP6 for osteoblast differentiation, (2) we will study BMP6 signaling that directs osteoblast differentiation, and (3) we will study the importance of BMP6-induced expression of osterix. These aims will be pursued in vitro, in vivo, and computationally by a team of investigators that includes a veterinary scientist and an orthopaedic surgeon - both trained as biochemists - and a computational biologist. Mechanistic experiments examining BMP6 function and cell signaling will be pursued in vitro and will utilize primary hMSC and human embryonic stem cells (lines WA01 and UC06). In vivo studies will use a preclinical heterotopic bone implantation model to examine the in vivo bone forming capacity of hMSC. Mathematical modeling will be used to describe kinetics of BMP6-induced osteoblast differentiation, and Bayesian learning will be used to discover signaling networks that are unique to BMP6-induced osteoblast differentiation. The completion of this five-year project will not only demonstrate the utility of BMP6 therapeutically to control hMSC osteoblast differentiation, but will also reveal novel signaling and transcriptional networks that govern hMC osteoblast differentiation. Long term, the identity of novel targets will lead to the development of enhanced therapeutic options for directing hMSC differentiation in tissue engineered constructs to repair bone.